Date Available


Year of Publication


Degree Name

Doctor of Philosophy (PhD)

Document Type

Doctoral Dissertation


Public Health



First Advisor

Dr. John Watkins

Second Advisor

Dr. David Fardo


There is evidence that cholesterol presents an important risk factor for Alzheimer’s disease (AD), but the direction of this relationship is modified by age. High cholesterol during midlife and low cholesterol during late life are both associated with an increased risk for AD. This dissertation research engaged a life span approach to study the relationship between cholesterol, AD and cognitive functioning among older adults. The purpose of this research was to determine if trajectories of cholesterol from midlife through late life differ according to AD status and if these trajectories are associated with cognitive functioning during old age.

This research employed a secondary analysis of cognitive, phenotypic and genetic data collected from subjects of the Framingham Heart Study (FHS) Original and Offspring Cohorts. Aim One involved creating three summary scores of the FHS neuropsychological battery. ROC analysis was used to determine which score best differentiated between cognitively normal, impaired and dementia subjects. Aim Two used generalized additive mixed models to examine trajectories of total, HDL and total/HDL cholesterol ratio according to AD status in the Original Cohort. Aim Three used mixed-effects models to examine the relationship between subject-specific trajectories of total cholesterol and cognition during old age.

Aim One determined that a summary score that provided equal weight to each assessment in the FHS neuropsychological battery best differentiates between subjects classified as cognitively normal, cognitively impaired and dementia. The findings from Aim Two indicated that there are subtle differences in the longitudinal trajectories of total, HDL and total/HDL ratio according to AD status. The findings from Aim Three provide limited evidence for a relationship between subject-specific trajectories of total cholesterol and cognitive functioning later in life. Older adults in the highest quartile for cognitive functioning had lower total cholesterol at approximately 55 years of age, but there were no differences in the mean trajectories of total cholesterol according to cognitive functioning later in life.

The findings from this research suggest that older adults with high cognitive functioning have lower total cholesterol during midlife, but life span cholesterol trajectories do not appear to be associated with AD status or cognitive function.